Data report - Gapminder dataset

Last update: 2024-04-30 11:50:30

Code

library(tidyverse)
library(gapminder)
library(gt)
library(gtsummary)
library(ggrepel)  
library(plotly)
library(sjPlot)

This report uses the Gapminder dataset, which includes real-world country-level panel data (1952-2007) for:

• Life expectancy (lifeExp)

• Population size (pop)

• GDP percapita (gdpPercap)

The dataset is available from gapminder r package. For more information on gapminder see: https://www.gapminder.org/

Code

gapminder %>% 
  glimpse()

Rows: 1,704
Columns: 6
$ country   <fct> "Afghanistan", "Afghanistan", "Afghanistan", "Afghanistan", …
$ continent <fct> Asia, Asia, Asia, Asia, Asia, Asia, Asia, Asia, Asia, Asia, …
$ year      <int> 1952, 1957, 1962, 1967, 1972, 1977, 1982, 1987, 1992, 1997, …
$ lifeExp   <dbl> 28.801, 30.332, 31.997, 34.020, 36.088, 38.438, 39.854, 40.8…
$ pop       <int> 8425333, 9240934, 10267083, 11537966, 13079460, 14880372, 12…
$ gdpPercap <dbl> 779.4453, 820.8530, 853.1007, 836.1971, 739.9811, 786.1134, …

The following table summarizes the main variables across the years.

Code

gapminder %>% 
  select(-country) %>% 
  tbl_summary(by = year)

Characteristic	1952, N = 1421	1957, N = 1421	1962, N = 1421	1967, N = 1421	1972, N = 1421	1977, N = 1421	1982, N = 1421	1987, N = 1421	1992, N = 1421	1997, N = 1421	2002, N = 1421	2007, N = 1421
continent
Africa	52 (37%)	52 (37%)	52 (37%)	52 (37%)	52 (37%)	52 (37%)	52 (37%)	52 (37%)	52 (37%)	52 (37%)	52 (37%)	52 (37%)
Americas	25 (18%)	25 (18%)	25 (18%)	25 (18%)	25 (18%)	25 (18%)	25 (18%)	25 (18%)	25 (18%)	25 (18%)	25 (18%)	25 (18%)
Asia	33 (23%)	33 (23%)	33 (23%)	33 (23%)	33 (23%)	33 (23%)	33 (23%)	33 (23%)	33 (23%)	33 (23%)	33 (23%)	33 (23%)
Europe	30 (21%)	30 (21%)	30 (21%)	30 (21%)	30 (21%)	30 (21%)	30 (21%)	30 (21%)	30 (21%)	30 (21%)	30 (21%)	30 (21%)
Oceania	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)	2 (1.4%)
lifeExp	45 (39, 60)	48 (41, 63)	51 (43, 65)	54 (46, 67)	57 (49, 69)	60 (50, 70)	62 (53, 71)	66 (55, 72)	68 (56, 73)	69 (56, 74)	71 (56, 75)	72 (57, 76)
pop	3,943,953 (1,452,026, 9,168,198)	4,282,942 (1,568,811, 9,817,598)	4,686,040 (1,784,362, 10,980,084)	5,170,176 (2,034,768, 12,614,585)	5,877,996 (2,351,192, 14,679,200)	6,404,036 (2,759,717, 16,670,227)	7,007,320 (3,006,286, 18,407,325)	7,774,862 (3,194,990, 20,947,542)	8,688,686 (3,605,992, 22,705,382)	9,735,064 (3,770,150, 24,311,370)	10,372,918 (4,173,506, 26,545,556)	10,517,531 (4,508,034, 31,210,042)
gdpPercap	1,969 (865, 3,913)	2,173 (931, 4,876)	2,335 (1,059, 5,709)	2,678 (1,151, 7,076)	3,339 (1,257, 9,509)	3,799 (1,357, 11,204)	4,216 (1,363, 12,348)	4,280 (1,327, 11,994)	4,386 (1,271, 10,684)	4,782 (1,367, 12,023)	5,320 (1,410, 13,360)	6,124 (1,625, 18,009)
1 n (%); Median (IQR)

Life expectancy

Code

gapminder %>% 
      ggplot(aes(x=factor(year), y=lifeExp)) +
  geom_boxplot()+
stat_summary(fun.y = "mean", geom = "point",color="lightblue",size=3)+
  theme_classic() +
  labs(title = "Life expectancy increase over time",
       x = "",
       y = "Country life expectancy",
       caption = "Source:Gapminder") 

Code

gapminder %>% 
  filter(year == 2007) %>% 
      ggplot(aes(x=lifeExp)) +
  geom_histogram(bins=12,fill="lightblue",color="white")+
  theme_classic() +
  guides(color = FALSE) +
  labs(title = "Life expectancy variation in 2007",
       x = "Country life expectancy",
       y = "Frequency",
       caption = "Source:Gapminder") 

Code

gapminder %>% 
      ggplot(aes(x=lifeExp)) +
  geom_histogram(bins=12,fill="lightblue",color="white")+
  theme_classic() +
  facet_wrap(vars(year))+
  guides(color = FALSE) +
  labs(title = "Life expectancy variation over time",
       x = "Country life expectancy",
       y = "Frequency",
       caption = "Source:Gapminder") 

Code

gapminder %>% 
      ggplot(aes(x=continent, y=lifeExp)) +
  geom_boxplot()+
stat_summary(fun.y = "mean", geom = "point",color="lightblue",size=3)+
  theme_classic() +
  labs(title = "Life expectancy across continents in 2007",
       x = "",
       y = "Country life expectancy",
       caption = "Source:Gapminder") 

Code

gapminder %>% 
  mutate(country_israel = ifelse(country=="Israel","Israel","")) %>% 
ggplot(aes(x=year, y=lifeExp, group=country, color = country_israel,size = country_israel)) +
    geom_line(alpha=0.5)+
  scale_colour_manual(name = "", values=c("lightblue","black"))+
    scale_size_manual(name = "", values=c(0.2,2))+
    theme_classic()+  
  labs(title = "Life expectancy increase across countries",
       x = "",
       y = "Country life expectancy",
       caption = "Source:Gapminder") 

Code

gapminder_lifeExp_tab <- gapminder %>%
    group_by(continent, year) %>%
    summarise(lifeExp=median(lifeExp))


gapminder_lifeExp_tab %>%
    ggplot(aes(x=year, y=lifeExp, color=continent)) +
     geom_line(size=1) + 
     geom_point(size=1.5) +
    geom_label_repel(data = gapminder_lifeExp_tab %>% filter(year == 2007), aes(label = continent),
                   nudge_x = 1,
                   nudge_y = 1) +
  theme_classic() +
  guides(color = FALSE) +
  labs(title = "Life expectancy increase across continents",
       x = "",
       y = "Country median life expectancy",
       caption = "Source:Gapminder") 

GDP per capita

Code

gapminder %>% 
      ggplot(aes(x=factor(year), y=gdpPercap)) +
  geom_boxplot()+
stat_summary(fun.y = "mean", geom = "point",color="lightblue",size=3)+
  theme_classic() +
  labs(title = "GPD per capaita increase over time",
       x = "",
       y = "Country GPD per capaita",
       caption = "Source:Gapminder") 

You can see that there’s an outlier

Code

gapminder %>% 
  filter(year == 2007) %>% 
      ggplot(aes(x=gdpPercap)) +
  geom_histogram(bins=12,fill="lightblue",color="white")+
  theme_classic() +
  guides(color = FALSE) +
  labs(title = "Variation in GDP per capita 2007",
       x = "Country GPD per capaita",
       y = "Frequency",
       caption = "Source:Gapminder") 

Code

gapminder %>% 
      ggplot(aes(x=gdpPercap)) +
  geom_histogram(bins=12,fill="lightblue",color="white")+
  theme_classic() +
  facet_wrap(vars(year))+
  xlim(0,60000)+
  guides(color = FALSE) +
  labs(title = "Variation in GDP per capita over time",
       x = "Country GPD per capaita",
       y = "Frequency",
       caption = "Source:Gapminder") 

Code

gapminder %>% 
  filter(year==2007) %>% 
      ggplot(aes(x=continent, y=gdpPercap)) +
  geom_boxplot()+
stat_summary(fun.y = "mean", geom = "point",color="lightblue",size=3)+
  theme_classic() +
  labs(title = "GDP per capita across continents in 2007",
       x = "",
       y = "Country GPD per capaita",
       caption = "Source:Gapminder") 

Code

gapminder %>% 
  mutate(country_israel = ifelse(country=="Israel","Israel","")) %>% 
ggplot(aes(x=year, y=gdpPercap, group=country, color = country_israel,size = country_israel)) +
    geom_line(alpha=0.7)+
  scale_colour_manual(name = "", values=c("lightblue","black"))+
    scale_size_manual(name = "", values=c(0.2,2))+
    theme_classic()+  
  labs(title = "GPD per capaita increase across countries",
       x = "",
       y = "Country GPD per capaita",
       caption = "Source:Gapminder") 

Code

gapminder_gdpPercap_tab <- gapminder %>%
    group_by(continent, year) %>%
    summarise(gdpPercap=median(gdpPercap))


gapminder_gdpPercap_tab %>%
    ggplot(aes(x=year, y=gdpPercap, color=continent)) +
     geom_line(size=1) + 
     geom_point(size=1.5) +
    geom_label_repel(data = gapminder_gdpPercap_tab %>% filter(year == 2007), aes(label = continent),
                   nudge_x = 1,
                   nudge_y = 1) +
  theme_classic() +
  guides(color = FALSE) +
  labs(title = "GPD per capaita increase across continents",
       x = "",
       y = "Country GPD per capaita",
       caption = "Source:Gapminder") 

Code

gapminder_gdpPercap_tab <- gapminder %>%
  filter(country %in% c("Israel","Lebanon","Greece","Turkey","Syria")) %>% 
    group_by(country, year) %>%
    summarise(gdpPercap=median(gdpPercap))


gapminder_gdpPercap_tab %>%
    ggplot(aes(x=year, y=gdpPercap, color=country)) +
     geom_line(size=1) + 
     geom_point(size=1.5) +
    geom_label_repel(data = gapminder_gdpPercap_tab %>% filter(year == 2007), aes(label = country),
                   nudge_x = 1,
                   nudge_y = 1) +
  theme_classic() +
  guides(color = FALSE) +
  labs(title = "GPD per capaita increase across continents",
       x = "",
       y = "Country GPD per capaita",
       caption = "Source:Gapminder") 

Link between life expectancy and GDP per capita

2007

Code

gapminder %>% 
  filter(year==2007) %>% 
  ggplot(aes(x=gdpPercap,y=lifeExp,color=continent,size=pop))+
  geom_point()+
    geom_text(aes(label=country),size=4,hjust=-0.1,vjust=-0.2,color="black",
            data = gapminder %>% filter(year==2007,country=="Israel"))+
    theme_classic()+
    guides(color = FALSE,size=FALSE) +
  labs(title = "Link between GDP per capita and life expectancy across continents",
       x = "GDP per capita",
       y = "Life expectancy",
       caption = "Source:Gapminder") 

Code

p1 <- gapminder %>% 
  filter(year==2007) %>% 
  ggplot(aes(x=gdpPercap,y=lifeExp,color=continent,size=pop, text = paste("country:", country)))+
  geom_point()+
    geom_text(aes(label=country),size=4,hjust=-0.1,vjust=-0.2,color="black",
            data = gapminder %>% filter(year==2007,country=="Israel"))+
    theme_classic()+
    guides(color = FALSE,size=FALSE) +
  labs(title = "Link between GDP per capita and life expectancy across continents",
       x = "GDP per capita",
       y = "Life expectancy",
       caption = "Source:Gapminder") 


ggplotly(p1)

Code

gapminder %>% 
  filter(year==2007) %>% 
  ggplot(aes(x=gdpPercap,y=lifeExp,color=continent,size=pop))+
  geom_point()+
  geom_smooth(aes(group=1))+
    geom_text(aes(label=country),size=4,hjust=-0.1,vjust=-0.2,color="black",
            data = gapminder %>% filter(year==2007,country=="Israel"))+
    theme_classic()+
    guides(color = FALSE,size=FALSE) +
  labs(title = "Link between GDP per capita and life expectancy across continents",
       x = "GDP per capita",
       y = "Life expectancy",
       caption = "Source:Gapminder") 

Code

gapminder %>% 
  filter(year==2007) %>% 
  ggplot(aes(x=log(gdpPercap),y=lifeExp,color=continent,size=pop))+
  geom_point()+
  geom_smooth(method="lm")+
    geom_text(aes(label=country),size=4,hjust=-0.1,vjust=-0.2,color="black",
            data = gapminder %>% filter(year==2007,country=="Israel"))+
    theme_classic()+
    guides(color = FALSE,size=FALSE) +
  labs(title = "Link between GDP per capita and life expectancy across continents",
       x = "GDP per capita (log)",
       y = "Life expectancy",
       caption = "Source:Gapminder") 

Regression model

Code

tmod1 <- lm(lifeExp ~ log(gdpPercap) + year + continent, data=gapminder)


tab_model(tmod1)

	life Exp
Predictors	Estimates	CI	p
(Intercept)	-465.87	-498.57 – -433.17	<0.001
gdpPercap [log]	5.02	4.71 – 5.34	<0.001
year	0.24	0.22 – 0.26	<0.001
continent [Americas]	8.93	8.02 – 9.83	<0.001
continent [Asia]	7.06	6.29 – 7.84	<0.001
continent [Europe]	12.51	11.51 – 13.51	<0.001
continent [Oceania]	12.75	10.25 – 15.25	<0.001
Observations	1704
R2 / R2 adjusted	0.798 / 0.797

Code

plot_model(tmod1)

Code

plot_model(tmod1,type = "std")

Code

plot_model(tmod1,type = "pred")

$gdpPercap

$year

$continent

Code

tmod2 <- tmod1 %>% update(.~. + continent* (log(gdpPercap) + year))


tab_model(tmod1,tmod2)

	life Exp			life Exp
Predictors	Estimates	CI	p	Estimates	CI	p
(Intercept)	-465.87	-498.57 – -433.17	<0.001	-478.42	-529.18 – -427.66	<0.001
gdpPercap [log]	5.02	4.71 – 5.34	<0.001	4.85	4.33 – 5.37	<0.001
year	0.24	0.22 – 0.26	<0.001	0.25	0.22 – 0.27	<0.001
continent [Americas]	8.93	8.02 – 9.83	<0.001	-40.71	-131.67 – 50.25	0.380
continent [Asia]	7.06	6.29 – 7.84	<0.001	-163.80	-247.02 – -80.59	<0.001
continent [Europe]	12.51	11.51 – 13.51	<0.001	306.46	214.37 – 398.55	<0.001
continent [Oceania]	12.75	10.25 – 15.25	<0.001	221.30	-465.04 – 907.63	0.527
gdpPercap [log] × continent [Americas]				2.46	1.33 – 3.59	<0.001
gdpPercap [log] × continent [Asia]				-0.02	-0.71 – 0.66	0.948
gdpPercap [log] × continent [Europe]				0.09	-1.01 – 1.20	0.867
gdpPercap [log] × continent [Oceania]				-1.61	-25.81 – 22.58	0.896
year × continent [Americas]				0.01	-0.03 – 0.06	0.547
year × continent [Asia]				0.09	0.04 – 0.13	<0.001
year × continent [Europe]				-0.15	-0.20 – -0.10	<0.001
year × continent [Oceania]				-0.10	-0.56 – 0.36	0.678
Observations	1704			1704
R2 / R2 adjusted	0.798 / 0.797			0.813 / 0.811

Code

plot_model(tmod2,type = "pred", terms = c("year","continent"))

Code

anova(tmod1,tmod2)

Analysis of Variance Table

Model 1: lifeExp ~ log(gdpPercap) + year + continent
Model 2: lifeExp ~ log(gdpPercap) + year + continent + log(gdpPercap):continent + 
    year:continent
  Res.Df   RSS Df Sum of Sq      F    Pr(>F)    
1   1697 57348                                  
2   1689 53159  8    4189.6 16.639 < 2.2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1